How are mothers and fathers balancing work and family under lockdown? - IFS Briefing Note BN290
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
How are mothers and fathers balancing work and family under lockdown? IFS Briefing Note BN290 Alison Andrew Sarah Cattan Monica Costa Dias Christine Farquharson Lucy Kraftman Sonya Krutikova Angus Phimister Almudena Sevilla
How are mothers and fathers balancing work and family under lockdown? Alison Andrew Sarah Cattan Monica Costa Dias Christine Farquharson Lucy Kraftman Sonya Krutikova Angus Phimister Almudena Sevilla Copy-edited by Judith Payne Published by The Institute for Fiscal Studies, May 2020 ISBN 978-1-912805-80-8 The authors are grateful to the Nuffield Foundation for funding this work (grant EDO/FR- 000022584). The Nuffield Foundation is an independent charitable trust with a mission to advance educational opportunity and social well-being. It funds research that informs social policy, primarily in Education, Welfare and Justice. It also provides opportunities for young people to develop skills and confidence in science and research. The Foundation is the founder and co-funder of the Nuffield Council on Bioethics, the Nuffield Family Justice Observatory and the Ada Lovelace Institute. Almudena Sevilla thanks the European Research Council for funding her time through the PARENTIME project. Co-funding from the ESRC-funded Centre for the Microeconomic Analysis of Public Policy (ES/M010147/1) is gratefully acknowledged by IFS researchers.
Executive summary
The COVID-19 crisis has caused drastic changes to most parents’ work lives and other
responsibilities. Millions of adults have lost or are forecast to lose their jobs permanently;
many more have stopped work temporarily. Others are newly working from home, while
many key workers are experiencing additional pressures and risks in their work. For most
parents, school and childcare closures have meant that children are at home, and
requiring care, for at least an extra six hours a day.
Although an economic downturn may be inevitable, the effects of COVID-19 and the public
health response on the economy as a whole and on specific groups are likely to look very
different from those of the economic crises that we are used to. The complete shutdown
of certain sectors and the huge increase in households’ care responsibilities are both
completely new. And the impacts of these sectoral shutdowns and increases in family
responsibilities are likely to look very different for mothers and fathers.
Prior to the crisis hitting, the shut-down sectors disproportionately employed women.1
And prior to the crisis, mothers typically performed a larger share of childcare and
housework than fathers did. This means that, if households continue to divide up
domestic responsibilities as they did before, mothers will take on a greater share of the
new responsibilities at home and will see a bigger absolute increase in the time they
spend on childcare and housework. On the other hand, such a substantial shock to
families’ typical arrangements could reshape how families divide paid work and unpaid
household responsibilities, with many fathers now at home all day with more exposure to
the scale and scope of housework and childcare.
The way that couples divide paid work and household responsibilities during this crisis
could have an effect that lasts long after the lockdown is lifted. If, on average, mothers are
more likely to step back from paid work during this crisis (either voluntarily or through
temporary or permanent job loss) and are more likely to pick up more of the domestic
responsibilities, they could face a long-run hit to their earnings prospects. This risks
reversing some of the progress that has been made on closing the gender wage gap.
On the other hand, previous evidence suggests that fathers who pick up more household
responsibilities (such as childcare) for a limited period of time may do a greater share of
them in the longer term.2 The work culture that relies on the ability of men to do paid work
being unaffected by fatherhood may change quickly over this period too, as employers
and co-workers may adapt to male employees requiring more flexibility to balance paid
work with childcare.
In this report, we examine new data collected since the end of April to investigate how this
crisis has affected mothers and fathers in two-parent opposite-gender families. Our
1
R. Joyce and X. Xu, ‘Sector shutdowns during the coronavirus crisis: which workers are most exposed?’, IFS
Briefing Note BN278, 2020, https://www.ifs.org.uk/publications/14791.
2
See A. Patnaik, ‘Reserving time for Daddy: the consequences of fathers’ quotas’, Journal of Labor Economics,
2019, 37, 1009–59, https://doi.org/10.1086/703115, M. Tamm, ‘Fathers’ parental leave-taking, childcare
involvement and labor market participation’, Labour Economics, 2019, 59, 184–97,
https://doi.org/10.1016/j.labeco.2019.04.007 and L. Farré and L. González, ‘Does paternity leave reduce
fertility?’, Journal of Public Economics, 2019, 172, 52–66, https://doi.org/10.1016/j.jpubeco.2018.12.002.
2 © Institute for Fiscal Studiessurvey allows up-to-date insights into the labour market shocks that parents have faced
and how parents are balancing their responsibilities under lockdown. In this piece, we
focus on how parents divide their time between childcare, housework and paid work, how
these arrangements are changing with the lockdown, and the extent to which they vary
between men and women.3
Key findings
All points relate to mothers and fathers in households with two opposite-gender
parents. Hours of time refer to the number of hour-long slots in which individuals
reported doing a given activity on a weekday.
The pressure on parents’ time is immense. On average, parents are doing
childcare during 9 hours of the day, and housework during 3. Among all
parents, paid work now takes up an average of just 3 hours, partly driven by
the large losses in employment. By contrast, comparable figures from 2014/15
suggest that, on a regular school day, parents did 5½ hours of childcare and
6½ hours of paid work. Furthermore, parents are now often doing at least two
activities at the same time.
Mothers are more likely to have quit or lost their job, or to have been furloughed,
since the start of the lockdown. Of parents who were in paid work prior to the
lockdown, mothers are one-and-a-half times more likely than fathers to have
either lost their job or quit since the lockdown began. They are also more
likely to have been furloughed. In all, mothers who were in paid work in
February are 9 percentage points less likely to be currently working for pay
(either remotely or on-site) than fathers.
Compared with fathers, mothers are spending less time on paid work but
more time on household responsibilities. The time they spend on paid
work is also more likely to be interrupted with household responsibilities.
Mothers are doing paid work during 2 fewer hours of the day than fathers,
but they do childcare and housework during 2 more hours each. Mothers
combine paid work with other activities (almost always childcare) in 47% of
their work hours, compared with 30% of fathers’ work hours.
The differences in work patterns between mothers and fathers have grown
since before the crisis. In 2014/15, mothers were in paid work at 80% of
the rate of fathers; now this is 70% of the fathers’ rate. Mothers in paid
work used to work an average of 73% of the hours that fathers worked;
this has fallen to 68%. Mothers and fathers used to be interrupted during
the same proportion of their work hours; now mothers are interrupted
over 50% more often.
3
Single parents face different challenges around how to meet, often on their own, the increased childcare and
housework responsibilities the crisis has created. Studying single parents raises specific issues that we do not
address in this report.
© Institute for Fiscal Studies 3Overall, in 2014/15, the average mother (including those who did not work
for pay) was doing nearly 60% of the number of uninterrupted work hours
that the average father did; now she is doing only 35%. Since difficult
labour market conditions are likely to persist for some time, reductions in
paid work now may also persist even after the immediate health crisis is
diminished. Workers who have lost their jobs permanently may struggle
to find new ones, workers who have reduced their hours may struggle to
increase them again, and workers whose productivity has suffered due to
interruptions may be penalised in pay and promotion decisions. The
disproportionate decrease in mothers’ paid work now suggests any long-
lasting effects will be particularly severe for them.
Families respond differently to a partner stopping paid work depending on
whether it is the mother or the father who stops. Mothers who have
stopped working for pay during lockdown while their partner continues
do twice as much childcare and housework as their partner. In the reverse
situation, in families where the father has stopped working, the parents
share childcare and housework equally, while the mother also does 5
hours of paid work a day.
The gaps between mothers’ and fathers’ time use are not straightforwardly
explained by mothers’ lower employment rates or lower earnings. Gender
gaps in time use remain even when comparing mothers working for pay with
fathers working for pay and when comparing mothers and fathers not
currently in paid work. Likewise, even in families where mothers were the
higher earner before the crisis and both partners are still working, mothers
still do more childcare and the same amount of housework as their partner.
Despite doing less childcare than mothers, during lockdown fathers have
nearly doubled the time they spend on childcare. On average, fathers are
now doing some childcare during 8 hours of the day, compared with 4 hours in
2014/15. This increase is especially large for the 15% of fathers in previously
dual-earner households who have lost their job while their partner continues
to do paid work. This large increase in fathers’ involvement in childcare might
have long-lasting impacts on how couples share childcare responsibilities.
4 © Institute for Fiscal Studies1. Introduction
The COVID-19 crisis has forced dramatic shifts in how families with children allocate their
time. With schools closed, most childcare off the table and very limited opportunities even
to leave the house, many families are having to maintain a difficult balancing act,
combining paid work, housework and full-time childcare. How are mothers and fathers
navigating these challenges, and what impact is the crisis having on gender inequalities?
To help answer some of these questions, researchers from the Institute for Fiscal Studies
(IFS) and the Institute of Education (IoE), with funding from the Nuffield Foundation, have
surveyed families with children in England, asking about how parents and children are
spending their time during lockdown and how their economic activity has changed. Box 1
describes these survey data.4
In this report, we summarise the initial evidence on how mothers and fathers living in
households with two opposite-gender parents in England were using their time at the end
of April and early May, over a month into lockdown and when only a limited number of
workers were able to leave the house to work. These statistics provide a clear, real-time
picture of how parents allocate responsibilities for childcare, housework and paid work
within the couple, how these arrangements changed with the lockdown, and whether they
depend on the relative income or educational attainment of partners.
While the group we focus on – parents in opposite-gender partnerships – include the
majority of parents, this work is not necessarily representative of how parents as a whole
are experiencing lockdown. The analysis does not include single parents, who make up
14.8% of families in England.5 Single parents are likely to face particular challenges due to
even more-exacerbated time pressures.6 Since most single-parent households are headed
by women, this is particularly important for how the crisis will affect mothers and fathers
differently. Likewise, the analysis does not include households with two parents of the
same gender, who typically divide up responsibilities differently from opposite-gender
parents.7 Both groups merit specific and careful analysis.8
4
These data have also been featured in a briefing note on home learning during the lockdown: A. Andrew, S.
Cattan, M. Costa Dias, C. Farquharson, L. Kraftman, S. Krutikova, A. Phimister and A. Sevilla, ‘Learning during
the lockdown: real-time data on children’s experiences during home learning’, IFS Briefing Note BN288, 2020,
https://www.ifs.org.uk/publications/14848.
5
https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/families/bulletins/
familiesandhouseholds/2019.
6
S. Cattan, C. Farquharson, S. Krutikova, A. Phimister and A. Sevilla, ‘Trying times: how might the lockdown
change time use in families?’, IFS Briefing Note BN284, 2020, https://www.ifs.org.uk/publications/14814.
7
M. E. Andresen and E. Nix, ‘What causes the child penalty? Evidence from same sex couples and policy
reforms’, Statistics Norway Research Department, Discussion Paper 902,
https://ideas.repec.org/p/ssb/dispap/902.html.
8
While the size of this data set is sufficient to carry out a detailed analysis of how opposite-gender two-parent
households (the majority group) are experiencing the crisis, the size limits our ability to conduct specific
analysis of these two important groups that addresses the particular challenges each may be facing. As larger
data sets collected during the crisis are released (for example, the Understanding Society panel), more
analysis of these groups will be possible.
© Institute for Fiscal Studies 5Box 1. IFS–IoE survey of families’ time use
Sample: Between 29 April and 15 May 2020, researchers from IFS and the IoE surveyed
4,915 parents who currently live with their children in England through an online survey.
We included families with children in eight different school years aged between 4 and
15.a In this report, we focus on opposite-gender two-parent households, of which we
have 3,591 in the data. We ensured that respondents were diverse in terms of their
gender, education, region, marital status, work status and the job they did. We then
reweighted our data to ensure that our statistics are representative of parents in
opposite-gender partnerships in England as a whole (see the appendix for details). In
particular, we were careful to reweight by how exposed parents’ jobs are to the crisis,
both in terms of whether the sector they work for has been shut down due to the
lockdown and in terms of whether home working is possible in their job.
Data collected: The main aim of our survey was to collect detailed information on how
families and children spend their time on a term-time weekday. We asked the surveyed
parent and their partner to fill in an online time-use diary, telling us what activities they
did during each hour of the day. We also asked the surveyed parent to fill in a similar
diary about their child’s time use (selecting one child at random in multi-child families),
and asked who the child was with during each time slot. Finally, we collected information
about the types of home learning activities children are doing, what resources have been
provided by the school and what resources are available at home for learning.
Interpreting time-use data: In order to keep the survey a manageable length for
families, we asked about time use in one-hour slots. Since these are wider than the 10-
minute intervals used in the most detailed time-use surveys, such as the 2015 UK Time
Use Survey, we cannot say precisely how long respondents spent on a particular activity;
respondents could report multiple activities during the hour, so the apparent number of
hours might overstate how long the respondent spent on the activities in that category.
Instead, we comment on the number of one-hour slots during which doing at least some
of a particular activity is reported.
a
We interviewed parents with children entering Reception next year and those with children in school in
Reception and in Years 1, 4, 5, 8, 9 and 10.
Several of our findings suggest that there is a risk of reversing some of the progress made
in narrowing the gender wage gap over recent decades. In line with other labour market
surveys collected since the start of the crisis,9 we find that women are more likely to have
stopped doing paid work since lockdown began. Further, even if they are doing some paid
work, women are doing fewer hours than men and are more likely to be juggling paid
work and childcare at the same time. Past research shows that the time women take off
when having a child, and the reduction in hours once they return to work, have long-term
effects, reducing their future hourly wages.10 There is a risk, therefore, that the differences
9
A. Adams-Prassl, T. Boneva, M. Golin and C. Rauh, ‘Inequality in the impact of the coronavirus shock: evidence
from real time surveys’, IZA, Discussion Paper 13183, 2020, http://ftp.iza.org/dp13183.pdf.
10
R. Blundell, M. Costa Dias, C. Meghir and J. Shaw, ‘Female labor supply, human capital, and welfare reform’,
Econometrica, 2016, 84, 1705–53, https://doi.org/10.3982/ECTA11576.
6 © Institute for Fiscal Studieswe find in how the work hours of fathers and mothers are being affected by this crisis will result in larger detrimental effects on the career progression and earnings prospects of mothers than of fathers. Also worthy of note, however, is the substantial increase in the time that the average father spends on childcare; our data suggest that the time the average father spends on childcare during a school day has doubled since before the crisis. Whether this change creates the impetus for longer-lasting change in how many parents share childcare is an important open question. © Institute for Fiscal Studies 7
2. How has the lockdown affected
parents’ paid work?
COVID-19 has brought an unprecedented disruption to working patterns, changing who is
in paid work and where, when and how they are working. Of the parents in our
(reweighted) sample who were doing some paid work during February 2020, only 54%
were still engaging in paid work at the time of the survey. 13% were no longer working for
pay due to having lost their job permanently (through being laid off, being fired or
quitting), while another 32% of parents were no longer working for pay due to having
been furloughed. Box 2 summarises how these figures compare with other estimates of
furloughing and unemployment rates.
While it is early to say whether our figures on job loss and furlough among parents are
precise estimates of what is happening in the wider economy, in this report we focus on
the differences in employment and time use between fathers and mothers and across
families, which are likely to be less sensitive to potential sampling bias.
Box 2. Comparing different estimates of job loss and furloughing
So far, there are no conclusive statistics on the rate of furloughing or of job loss and the
statistics that do exist vary widely. Official estimates for the UK suggest that 23% of
those who were working for pay (employed or self-employed) before the crisis had been
furloughed by mid May.a However, a recent online survey of the labour market
suggested a much higher figure of 43%.b In all, the proportion of parents in opposite-
gender partnerships who were previously working that we estimate to have stopped
working due to having been furloughed (32%) lies between official figures and those
from recent labour market surveys.
Official information about how many people have lost their job is more scarce. A huge
increase in the volume of claims made for benefits related to unemployment and
financial hardship indicate the unprecedented scale of the financial challenges facing
households. During the first eight weeks after social distancing was announced (on 16
March), there were 2.6 million individual ‘declarations’ from people applying for
universal credit benefits.c Not all these claims will, however, relate to new job loss. The
government estimates that as of 9 April, three weeks into social distancing, 856,000 more
people than one month earlier were claiming benefits principally for the reason of
unemployment;d these numbers are likely to increase substantially as they are updated
to cover a greater period of the lockdown. They are also likely to underestimate the true
extent of job losses as not all job losses will have resulted in new benefit claims: some
newly unemployed workers will already have been claiming universal credit so will not
have needed to make a new claim, others will not have met the eligibility criteria, while
still more will have been entitled to claim but will not have actually done so.
8 © Institute for Fiscal StudiesThe difference in reference period (our data capture those who have stopped working
for pay up until mid May, a month later than the official estimates of new benefit claims
for the principal reason of unemployment) makes it difficult to compare our estimates
with these official statistics. So does the fact that we look at all those who have stopped
working for pay (due to being laid off, being fired or having quit) whereas the
government figures that are available so far only relate to those claiming benefits due to
unemployment, which is a narrower group. While it is difficult to make comparisons with
official figures, our estimate of the proportion of parents no longer working for pay (13%
of those who were previously working) is similar to, but slightly lower than, in a recent
labour market survey which estimated a figure of 15%.b
Importantly, our estimates of the prevalence of both furlough and stopping work for
other reasons are not directly comparable to either official statistics or recent labour
market surveys: we focus on opposite-gender, dual-parent households with dependent
children in England, rather than all workers in the UK. This could be particularly
important since parents are able to ask to be furloughed because of pandemic-related
caring responsibilities, including childcare while schools are closed. Since our figures
apply only to parents with dependent children, they are less useful as indicators of the
health (or lack thereof) of the UK labour market as a whole. However, they do clearly
indicate that parents with dependent children have seen enormous disruption to their
working patterns.
1
Official estimates report that 7.5 million (employee) jobs had been furloughed by 12 May
(https://www.gov.uk/government/news/chancellor-extends-furlough-scheme-until-october). This
compares with Office for National Statistics (ONS) estimates, based on the Labour Force Survey, that
there were 33.1 million in paid work in the UK between December 2019 and February 2020 (ONS Dataset
EMP01 SA, release date 19 May 2020).
b
A. Adams-Prassl, T. Boneva, M. Golin and C. Rauh, ‘Inequality in the impact of the coronavirus shock:
evidence from real time surveys’, IZA, Discussion Paper 13183, 2020, http://ftp.iza.org/dp13183.pdf.
c
Department for Work and Pensions, ‘Universal Credit declarations (claims) and advances: management
information’, released 19 May 2020, https://www.gov.uk/government/publications/universal-credit-
declarations-claims-and-advances-management-information#history.
d
Office for National Statistics, ‘Employment in the UK: May 2020’, released 19 May 2020,
https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/b
ulletins/employmentintheuk/may2020.
The adverse effects of this crisis may not impact mothers’ and fathers’ ability to do paid
work equally. First, mothers are more likely than fathers to work in the sectors that are
taking the biggest hit from the lockdown.11 This aspect is different from in previous
recessions, in which male-dominated sectors suffered the most.12
11
T. Alon, M. Doepke, J. Olmstead-Rumsey and M. Tertilt, 2020, ‘The impact of Covid-19 on gender equality’,
Covid Economics: Vetted and Real-Time Papers, 4, 62–85,
https://cepr.org/sites/default/files/news/CovidEconomics4.pdf. Also R. Joyce and X. Xu, ‘Sector shutdowns
during the coronavirus crisis: which workers are most exposed?’, IFS Briefing Note BN278, 2020,
https://www.ifs.org.uk/publications/14791.
12
M. Doepke and M. Tertilt, ‘Families in macroeconomics’, ch. 23 in J. B. Taylor and H. Uhlig (eds), Handbook of
Macroeconomics: Volume 2, North-Holland, Amsterdam, 2016.
© Institute for Fiscal Studies 9Moreover, the COVID-19 crisis has been distinguished by the sudden, near-total loss of
access to schools and childcare, leaving parents with enormous additional responsibilities
for childcare and education. Since mothers already spent more time on childcare and
other unpaid work,13 and since in many couples the woman is the lower earner, it is
possible that these additional responsibilities are being disproportionately shouldered by
mothers.
We find important differences in the rates of job loss and of furloughing between mothers
and fathers. We see in Figure 1 that, among parents who were working in February 2020,
mothers are indeed 9 percentage points more likely to have stopped working for pay than
fathers. 16% of mothers are no longer doing paid work due to having lost their work
permanently (whether they were laid off, were fired or quit), compared with 11% of
fathers. Mothers are also somewhat more likely to not be doing paid work due to having
been furloughed through the Coronavirus Job Retention Scheme (34%, compared with
30% of fathers). These effects compound the already unequal employment rates of
mothers and fathers, which in our data were respectively 75% and 92% in February 2020
(very close to the 75% and 93% in nationally representative data for April to June 2019).14
So, while prior to the crisis mothers were in paid work at 80% of the rate that fathers were,
now they are in paid work at only 70% of the rate.
Figure 1. Current engagement in paid work by gender for parents who were in paid
work in February 2020
Mothers
Fathers
0% 20% 40% 60% 80% 100%
Working outside the home Working from home
Not working: furloughed Not working: laid off, fired or quit
13
S. Cattan, C. Farquharson, S. Krutikova, A. Phimister and A. Sevilla, ‘Trying times: how might the lockdown
change time use in families?’, IFS Briefing Note BN284, 2020, https://www.ifs.org.uk/publications/14814.
14
See ONS, ‘Families and the labour market, UK: 2019’,
https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/article
s/familiesandthelabourmarketengland/2019. These employment rates are also consistent with findings from
other surveys recently collected for the UK (e.g. A. Adams-Prassl, T. Boneva, M. Golin and C. Rauh, ‘Inequality
in the impact of the coronavirus shock: evidence from real time surveys’, IZA, Discussion Paper 13183, 2020,
http://ftp.iza.org/dp13183.pdf; A. Sevilla and S. Smith, ‘Baby steps: the gender division of childcare during the
COVID19 pandemic’, Centre for Economic Policy Research (CEPR), Discussion Paper 14804, 2020,
https://cepr.org/active/publications/discussion_papers/dp.php?dpno=14804).
10 © Institute for Fiscal StudiesIf those who have stopped paid work during the crisis find it difficult to return in the short
term, either because low labour demand coupled with high overall levels of
unemployment makes it hard to find a job or because their other commitments remain
incompatible with paid work, these initial inequalities could persist beyond this crisis
through the loss of skills and labour market attachment leading to long-term increases in
gender inequalities.
Differences by parents’ education qualifications
The lockdown has also opened up inequalities between education groups, as Figure 2
highlights. Amongst both mothers and fathers, individuals with fewer qualifications are
more likely to have stopped paid work since the start of the crisis. As has been discussed
elsewhere,15 this may well be because more-skilled jobs can be done from home more
easily. For example, over three-quarters of university-educated parents in our data who
are currently working are working from home, compared with well under half of those
with GCSE qualifications or below.
The gender inequalities in who has stopped working are clear within all three education
groups: mothers are always more likely to have stopped working than fathers,
independent of their qualifications. Strikingly, the gender gap is similar at the top and the
bottom of the distribution of education: among degree-educated parents, mothers are 9.2
percentage points more likely to have stopped work, while the gap is 8.8 percentage
points among those educated to GCSE level or below. The gap for the group with A-level
qualifications is around two-thirds as big again.
Figure 2. Current work status by gender and education for parents who were in paid
work in February 2020
GCSE or below
Mothers
A Levels
Degree
GCSE or below
Fathers
A Levels
Degree
0% 20% 40% 60% 80% 100%
Working outside the home Working from home
Not working: furloughed Not working: laid off, fired or quit
15
M. Costa Dias, C. Farquharson, R. Griffith, R. Joyce and P. Levell, ‘Getting people back into work’, IFS Briefing
Note BN286, 2020, https://www.ifs.org.uk/publications/14829.
© Institute for Fiscal Studies 113. How are mothers and fathers
spending their time in lockdown?
Changes in employment patterns are not the only way in which the lockdown is impacting
how families spend their time; most children now rely on their parents as their sole
childcare and chief education providers. In this section, we document how mothers and
fathers are spending their time during lockdown and who is shouldering these additional
responsibilities.
In our survey, we asked the main respondent what activities they, their partner and one of
their children were doing during each hour of the previous weekday (the survey was
paused over weekends and Bank Holidays). The respondent could include more than one
activity in each hour slot. For parents, the activities we asked about were paid work,
housework, leisure, exercise, personal care, ‘active’ childcare (such as playing with a child
or doing educational activities), ‘passive’ childcare (keeping an eye on a child), caring for
others (not children) and sleep.16 In this analysis, we combine active and passive childcare
into one childcare category, we combine ‘exercise’ and ‘leisure’ to create one exercise
category and we drop ‘caring for others’ due to the rarity with which this activity was
selected. For children, in addition to age-appropriate activities, we asked who was
supervising them during each hour and we use this information to cross-check the hours
in which parents were doing passive childcare. This leaves us with six different categories
of activities during waking hours. In this section, we describe how, on average, all mothers
and fathers (regardless of whether they worked prior to the crisis and whether they are
working now) have been spending their time.
How are mothers and fathers spending their weekdays?
For each hour of the day, Figure 3 shows the share of mothers and fathers taking part in
each of these six categories of activity. In some cases, Figure 3 shows few differences
between men and women: their sleeping patterns, time spent on personal care and
leisure time look nearly identical.
But there are stark differences in time spent on paid work, housework and childcare. At all
points in the day, more men than women are doing paid work. For example, while around
a fifth of mothers report doing paid work between noon and 1pm, nearly two-fifths of
fathers say they are working then. The reverse is true for housework, with more women
doing housework during every hour of the day. In childcare too, the gender difference is
striking, particularly during core working hours; at noon, for example, around 70% of
mothers were doing childcare compared with 50% of fathers.
These differences mean that mothers report spending at least some of their time on
housework and childcare in more hour-long slots during the day than their partners do.
Figure 4 summarises these findings, showing the total number of one-hour slots in which
mothers and fathers report doing childcare, housework and paid work.
16
In addition to respondents’ reports of passive childcare done by them and their partner, we also include
hours in which the respondent reported through the child time-use diary that they or their partner was
supervising their child.
12 © Institute for Fiscal StudiesFigure 3. Mothers’ and fathers’ time use over the course of the day
Figure 4. Total time slots reported by mothers and fathers
Childcare
Housework
Paid work
0 1 2 3 4 5 6 7 8 9 10 11
One-hour time slots per day
Fathers Mothers
Averaging across those currently doing paid work and those who are not, Figure 4 shows
that, on average, fathers report doing some paid work in four one-hour slots, two more
than the average for mothers. On the other hand, fathers report doing housework in just
over two time slots, compared with nearly four for mothers. There is a similar story for
childcare, where mothers report doing childcare in over ten one-hour slots and fathers do
so in eight. Of course, one reason for these gender differences is that – as Section 2 shows
© Institute for Fiscal Studies 13– men are more likely to remain in paid work during the crisis and, indeed, are more likely
to have been working for pay, particularly in full-time work, prior to the crisis. We return to
this in further detail below.
Gender differences in childcare
The differences in time spent on childcare, shown in Figure 4, are stark: on average,
mothers are engaged in some childcare – whether active or passive – in over 25% more
hour-long slots than fathers. But even for fathers, childcare is the activity that is most
frequently reported during waking hours (see Figure 3).17 This means that parents,
especially mothers, are doing some childcare throughout the great majority of their day.
To put these patterns into context, we now look at how these figures compare with the
amount of childcare that parents were doing prior to the crisis, using data from the UK
Time Use Survey (UKTUS), a nationally representative time-use survey carried out in
2014/15.18
Figure 5 shows that, on an average school day in 2014/15, fathers were engaged in
childcare during 4.2 one-hour slots and mothers during 6.7 slots; the 8.0 and 10.3 one-
hour slots in which, on average, fathers and mothers report doing childcare over the last
few weeks are thus very substantial increases. The increase is particularly large for
fathers, who have nearly doubled the number of slots in which they engage in childcare.
Such a sudden and significant change may have longer-run effects on how parents share
childcare going forward, even after lockdown, and on how employers view male
employees’ childcare responsibilities.
Of course, not all childcare requires the same level of effort from the parent; it could be
much easier to combine cooking with keeping an eye on a child while they watch TV than
it is to combine writing work reports with playing Lego. Figure 5 therefore also breaks our
measure of childcare into ‘active’ and ‘passive’ care. It shows that more than half (56% for
mothers and 61% for fathers) of the time spent looking after children is taken up with
‘passive childcare’ – keeping an eye on the children or watching TV together, for example
– rather than ‘active childcare’, such as doing schoolwork or playing together.
17
This is consistent with other recent surveys which also find large differences in the time spent on childcare by
mothers and fathers during the current crisis (e.g. A. Adams-Prassl, T. Boneva, M. Golin and C. Rauh,
‘Inequality in the impact of the coronavirus shock: evidence from real time surveys’, IZA, Discussion Paper
13183, 2020, http://ftp.iza.org/dp13183.pdf; A. Sevilla and S. Smith, ‘Baby steps: the gender division of
childcare during the COVID19 pandemic’, Centre for Economic Policy Research (CEPR), Discussion Paper
14804, 2020, https://cepr.org/active/publications/discussion_papers/dp.php?dpno=14804).
18
We run this analysis on a subsample of households with children of similar ages to those in our survey
(focusing on children from age 8, when child time-use diaries are first available, to age 15). We also recode
data from the UKTUS to make them as comparable as possible to ours by recoding the survey’s 10-minute
intervals into hour-long intervals and recoding whether the respondent did any childcare during that hour.
Our measure of doing childcare in the UKTUS is based on cross-checking the reports of when children say
they were with parents against their parents’ diaries.
14 © Institute for Fiscal StudiesFigure 5. Active versus passive childcare and comparison with pre-lockdown time use
2014/15
Mothers
2020 April
2014/15
Fathers
2020 April
0 1 2 3 4 5 6 7 8 9 10 11
One-hour time slots per day
Active Passive Total
Note: 2014/15 figures use the UK Time Use Survey. This data set contains activity information down to 10-minute
slots of the day. We create a data set from this that indicates whether a parent did any childcare within a given
hour, to best mimic the methodology in our survey.
© Institute for Fiscal Studies 154. Quantity versus quality of work time
The amount of paid work that parents can do depends not just on how long they spend
working, but also on how productive they are during that time. Indeed, previous research
has shown that the total amount of working time and the total amount of focused,
uninterrupted working time are both important determinants of workers’ productivity and
learning.19
However, for parents working from home, working time – and especially focused working
time – can be hard to come by. Parents, who are now largely responsible for both
childcare and education around the clock, are contending with more demands from their
families on their time. And childcare, particularly for younger children, is often not an
activity that can be rigorously scheduled in focused blocks of time.
In this section, we explore what the lockdown has meant for parents’ working patterns.
Since we asked respondents to list all of the activities they did in each one-hour slot, our
data give us a clear picture of the extent to which parents’ hours of paid work are being
shared with – or interrupted by – other responsibilities. To understand how this crisis is
changing the quantity of working time, we compare current figures with what similar
parents used to do before the lockdown using the UK Time Use Survey.
Figure 6 illustrates how much working patterns have changed for fathers and mothers.
The top panel shows the average number of hour-long slots dedicated to paid work
before and during the crisis, for all mothers and all fathers taken together. Overall, we find
that fathers are doing paid work in fewer than half the number of hours that average
fathers used to work before the crisis. For mothers, the drop is even more staggering;
they are currently working in less than two-fifths of the number of hours that similar
mothers reported prior to the lockdown. In all, while in 2014/15 mothers who were doing
paid work were working 73% of the number of hours of fathers in paid work, now mothers
are working only 68% of fathers’ hours.
Of course, these figures are partly driven by the large changes in employment status that
we reported in Section 2. Panel B in Figure 6 shows how working hours among the
subgroup of parents who are working for pay during the lockdown period compared with
pre-lockdown figures for working parents. It shows that substantial gender differences
persist: while the working hours of fathers dropped by about 16%, those of mothers
dropped by 22%.
19
R. Blundell, M. Costa Dias, C. Meghir and J. Shaw, ‘Female labor supply, human capital, and welfare reform’,
Econometrica, 2016, 84, 1705–53, https://doi.org/10.3982/ECTA11576; A. Adams, ‘The gender wage gap in an
online labour market: the cost of interruptions’, Centre for Economic Policy Research (CEPR), Discussion Paper
DP14294, 2020, https://cepr.org/active/publications/discussion_papers/dp.php?dpno=14294; D. Coviello, A.
Ichino, N. Persico, ‘The inefficiency of worker time use’, Journal of the European Economic Association, 2015, 13,
906–47, https://doi.org/10.1111/jeea.12129.
16 © Institute for Fiscal StudiesFigure 6. Working hours before and during the lockdown
Panel A. Average working hours: all parents
2014/15
Mothers
2020 April
2014/15
Fathers
2020 April
0 1 2 3 4 5 6 7 8 9
One-hour time slots per day
Uninterrupted Interrupted
Panel B. Average working hours: working parents
2014/15
Mothers
2020 April
2014/15
Fathers
2020 April
0 1 2 3 4 5 6 7 8 9
One-hour time slots per day
Uninterrupted Interrupted
Note: 2014/15 figures use the UK Time Use Survey. This data set contains data down to 10-minute slots of the
day. We create a data set from this that indicates whether a parent did any paid work within a given hour, to best
mimic the methodology in our survey.
© Institute for Fiscal Studies 17Figure 6 also shows the amount of parents’ paid work hours that are multitasking (or are
interrupted), defined as doing at least one work and one non-work activity during the
hour-long slot. We see that, before the crisis, parents typically did non-work activities
during a quarter of the hours in which they were also doing paid work. Such hours most
typically occurred at the start and end of their workdays, as they transitioned to and from
work.20 We see that now, overall, parents who are currently working for pay (Figure 6B)
have a higher number of interrupted hours, both in absolute terms and as a fraction of
their total work hours. The nature of the interruptions is also different; during the
lockdown, in roughly 90% of interrupted work hours the parent is also doing childcare
alongside his or her paid work. Working time before the crisis was mostly spent in the
workplace, where interruptions are likely to be less frequent and less disturbing for the
workflow. The increase in interruptions is therefore unsurprising given many parents’
transition to doing their job from home with children out of school. The reduction in the
proportion of hours that parents work uninterrupted implies that the drop in the amount
of work that parents are able to do may be even more pronounced than suggested by the
drop in working hours. Our data confirm that indeed multitasking is less frequent among
those parents working outside the home, in which case it happens in 18% of work hours,
compared with 49% for those working from home.
Moreover, the extent of multitasking during work time is more prevalent among women
than men. While 70% of fathers’ work hours are spent exclusively doing work, this is the
case for only 53% of mothers’ work hours.21 In other words, mothers are being interrupted
during 57% more of their paid work hours than fathers. This was not the case before the
crisis: then, mothers and fathers were interrupted during the same proportion of their
work hours. Combined with the gaps in hours spent doing paid work, this amounts to the
average father who is currently working for pay having nearly twice as many
uninterrupted work hours as the average mother who is currently working for pay (5.1
hours versus 2.6 hours). This is a bigger gap in uninterrupted hours than we saw before
the crisis (6.6 hours for fathers versus 4.7 hours for mothers). Combined too with the fact
that mothers are more likely to have stopped working for pay since the lockdown and that
they were already less likely to be doing so before the crisis, mothers overall (including
those not working for pay at all) are working just over a third of the number of
uninterrupted hours that fathers are.
Taken together, we find that not only are mothers less likely to work during the lockdown,
but also, even if they are working, they spend fewer hours on paid work and the time they
do spend working is likely to be less productive than fathers’ work time because of
interruptions. Lower productivity during the time mothers spend on paid work – and
being paid – could itself impact their career prospects, making them seem less committed
to their jobs or less able to cope with their workloads than their male colleagues.
20
Commuting to and from paid work is included in our definition of hours spent doing some paid work.
21
These findings are unlikely to be driven by differences in reporting behaviour between men and women (e.g.
women being more likely to tick multiple boxes). Our analysis includes both the main respondent’s and
his/her partner’s time use. Since our main respondents are a mix of men and women, we have some mothers
responding about fathers and some fathers responding about mothers. We did not find systematic
differences in gender gaps for reported time use by gender of the main respondent.
18 © Institute for Fiscal Studies5. What drives differences in how
mothers and fathers spend their
time?
So far, we have seen that mothers and fathers are spending their time very differently in
the lockdown. While men have significantly increased the time they spend on childcare, it
is still mothers who are taking on the bulk of childcare and housework responsibilities,
often at the same time as they complete other tasks. By contrast, fathers are doing much
more paid work than mothers during the lockdown. These results suggest that the adults
in two-parent families are continuing – or intensifying – the specialisation they had already
developed before the pandemic, with one partner focusing more on paid work while the
other takes more responsibility for unpaid work at home.
Notably, this specialisation has been incentivised by policy during the pandemic: the
Coronavirus Job Retention Scheme means that couples where one parent stops doing paid
work entirely and is furloughed, while the other continues to work their regular hours, are
financially much better off than those where both partners reduce their working hours to
accommodate their new domestic responsibilities. But noting that there are incentives in
favour of specialisation within couples does not answer why, on average, specialisation
during the lockdown seems to split along gender lines, with mothers spending more time
on domestic responsibilities while fathers do more paid work. There could be a number of
reasons for this gendered division of labour, including:
Employment rates: As we have shown, fathers are more likely to be in paid
employment than mothers. This was already true before the crisis – 75% of mothers
were working between April and June 2019, compared with 93% of fathers22 – but the
differences are now even more pronounced.
Hourly earnings: On average, fathers earn more than mothers; before the crisis,
mothers whose first child was 5 years old earned more than 15% less per hour than
fathers with a first child the same age, and the gender wage gap continued to grow as
the child got older.23 This means that an extra hour that a father spends doing paid
work is, on average, more financially valuable for the household.
Productivity in domestic work: Similarly, if women are more capable or productive at
housework or childcare – for example, because they did more of it before the crisis and
so are more experienced – the household might benefit more from an extra hour of a
mother’s domestic work than it would from a father’s.
Preferences: The way in which partners share responsibility for childcare and
housework could also be influenced by partners’ preferences, habits or beliefs about
who ‘should’ take responsibility for these activities.
22
ONS, ‘Families and the labour market, UK: 2019’,
https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/article
s/familiesandthelabourmarketengland/2019 (accessed 21 May 2020).
23
M. Costa Dias, R. Joyce and F. Parodi, ‘Wage progression and the gender wage gap: the causal impact of hours
of work’, IFS Briefing Note BN223, 2018, https://www.ifs.org.uk/publications/10358.
© Institute for Fiscal Studies 19These are, of course, not the only reasons that could affect how mothers and fathers choose to divide their tasks; other influences, such as different accommodations from employers, could also play a role. Disentangling all of the different factors at play, and the extent to which each could influence how couples divide their responsibilities, is a complicated question and outside the scope of this report. But in this section, we explore the extent to which some of these causes appear to be at play in driving the gendered division of labour during lockdown. Are gender gaps explained by mothers being less likely to be employed? Even prior to the current crisis, mothers were less likely to be in paid work than fathers. On top of that, in this report we document that the lockdown has caused mothers to stop working for pay at higher rates than fathers. For equity and financial reasons, a spouse who is not working for pay may accumulate most of the domestic responsibilities within the family. Therefore, one explanation for the gender gaps we see in childcare and housework is simply that mothers are more likely to be not currently doing paid work. If this explanation were sufficient to explain the gender gaps we see, we would expect that mothers who are currently not doing paid work spend their time in the same ways as fathers who are not doing paid work. Panel A of Figure 7 shows that this is not the case. It graphs the number of one-hour slots in which parents report doing paid work, housework, and childcare (of any type), splitting between those who are currently working for pay and those – including furloughed employees – who are not. Panel B compares those who report they are working from home and those who report working outside of the home. In Figure 7A, we see large gender gaps even between mothers who are working for pay and fathers who are working for pay, and between mothers who are not currently working for pay and fathers who are not currently working for pay. Mothers spend more time than fathers on housework and childcare, regardless of working status. This tells us that the overall gender differences we observe in time spent on domestic activities are not entirely driven by fewer mothers currently doing paid work. Indeed, mothers who are working for pay spend roughly the same amount of time on these activities as do fathers who are not working for pay. Fathers’ time use is particularly sensitive to whether or not they are in paid work: fathers who are not working for pay report spending more than twice as many time slots on housework as those who are, and three more one-hour time slots on childcare. Interestingly, the amount of time that mothers spend on childcare is far less responsive to whether or not they work for pay. Indeed, mothers who are not working for pay do childcare in only one-and-a-half more one-hour time slots than those who are. Altogether this means that mothers who are working for pay have an especially heavy load. While they do 2.3 fewer one-hour time slots in paid work than working fathers, they more than compensate for the difference by putting in an additional 2.7 one-hour time slots of childcare and 1.5 one-hour time slots of housework compared with fathers in paid work. 20 © Institute for Fiscal Studies
Figure 7. Time use by current work status
Panel A. By working status
Not working
Childcare
Housework
Childcare
Working
Housework
Paid work
0 1 2 3 4 5 6 7 8 9 10 11
One-hour time slots per day
Fathers Mothers
Panel B. Among those working, by whether works at home or outside the home
Childcare
Outside
Housework
Paid work
Childcare
Home
Housework
Paid work
0 1 2 3 4 5 6 7 8 9 10 11
One-hour time slots per day
Fathers Mothers
© Institute for Fiscal Studies 21The large differences in time use we see between working mothers and fathers could be
driven by the type of paid work different genders are doing, which may or may not be
compatible with more time for domestic responsibilities. In particular, work outside the
home is better insulated from domestic responsibilities during working hours and, after
factoring in commuting time, it may also leave fewer hours available for these activities.24
To shed some light on the extent to which gender differences in domestic responsibilities
among working parents can be explained by the nature of their jobs, we split time use by
whether working parents work from home or outside of the home. Although both
mothers and fathers do more childcare if they are working from home, Figure 7B shows
that there are gender gaps both between mothers and fathers who work from home (with
mothers doing childcare during 2.4 more hours than men) and between parents working
outside the home (where the gap is even larger, at 3.2 one-hour slots).
Are gender gaps explained by mothers having lower earnings?
Before the pandemic, mothers earned less per hour on average than fathers. There is
evidence that they were more willing to sacrifice higher earnings in favour of, for example,
more flexible working or a shorter commute.25 This means that, in strict economic terms,
families on average benefit financially more from an extra hour of paid work done by a
man, even if it means his partner has to reduce her paid working hours to pick up more of
the responsibilities at home. We look now at whether the lower average earnings of
mothers relative to fathers can explain the overall gender gaps we see.
If this were the main explanation for the overall gender gaps, we would expect to see that
parents behave symmetrically. That is to say, we would expect that couples usually
prioritise the (paid) work of the higher-earning parent, regardless of whether that parent
is the mother or the father. This would imply that, on average, higher-earning partners do
the same amount of childcare, housework and paid work whether they are male or
female, and lower-earning fathers do the same amount as lower-earning mothers.
Again, this is not what we find. Figure 8 shows how partners who are both currently in
paid work are organising their time depending on who earned more before the crisis. It
shows clearly that couples where the father earned more do not organise their time in the
opposite way to which couples where the mother earned more do. Indeed, it shows that
regardless of who was the higher earner in the couple, the father always does more
uninterrupted work (2.8 hours if he earned more and 0.6 hours if the mother earned
more). Likewise, even when the mother earned more than her partner, she does 1.6 more
hours of childcare and about the same amount of housework.
24
Commuting to and from paid work is included in our definition of hours spent doing some paid work.
25
R. Joyce and A. Norris Keiller, ‘The “gender commuting gap” widens considerably in the first decade after
childbirth’, IFS Observation, 7 November 2018, https://www.ifs.org.uk/publications/13673. A. Mas and A.
Pallais, ‘Valuing alternative work arrangements’, American Economic Review, 2017, 107, 3722–59,
https://doi.org/10.1257/aer.20161500.
22 © Institute for Fiscal StudiesFigure 8. Time use by pre-crisis relative earnings
Childcare
Father earnt more
Housework
Paid work
Uninterrupted paid work
Childcare
Mother earnt more
Housework
Paid work
Uninterrupted paid work
0 1 2 3 4 5 6 7 8 9 10
One-hour time slots per day
Father Mother
Note: Earnings are 2019 pre-tax earnings for both partners. Sample includes only couples where both are
currently working.
Do gender gaps remain when we remove paid work from the picture?
The evidence above suggests that neither employment rates nor relative earnings can
fully explain the differences in how mothers and fathers are spending their time in
lockdown. This suggests that other considerations drive how parents share
responsibilities. One possible factor is parents’ preferences, habits or beliefs about the
best way to allocate responsibilities for paid and unpaid work. The dramatic changes in
parents’ work arrangements during the crisis provide a particularly interesting way of
examining whether this is the case.
We take families where both partners were in paid work before the crisis and look at what
happens when mothers stop working and when fathers stop working. Comparing what
happens when the father versus the mother stops work allows us to see whether there
are factors other than economic incentives at play in how parents divide up
responsibilities. If the drivers of the division of responsibilities were purely economic, we
would expect parents to behave symmetrically when either partner stops working: that
when the mother stops working but the father continues, the mother does most of the
domestic work and leaves her partner to focus on paid work, and vice versa.
© Institute for Fiscal Studies 23You can also read
